Electrochemical synthesis of polygermanes

Article abstract of DOI:10.1016/S0013-4686(98)00173-X

Five dialkyl-substituted and three phenyl substituted polygermanes were electrochemically synthesized. Among them, dibutyl-, dipentyl-, and dihexyl-substituted polygermanes gave the best synthetic results, i.e., relatively high yields (25-45%), high curre

Full text of DOI:10.1016/S0013-4686(98)00173-X

PERGAMON
Electrochimica Acta 44 (1998) 659±666
Electrochemical synthesis of polygermanes
Mitsutoshi Okano *, Ken-ichi Takeda, Takeshi Toriumi, Hiroshi Hamano
Department of Photo-Optical Engineering, Tokyo Institute of Polytechnics, 1583 Iiyama, Atsugi, Kanagawa 243-02, Japan
Received 15 December 1997; received in revised form 17 April 1998
Abstract
Five dialkyl-substituted and three phenyl substituted polygermanes were electrochemically synthesized. Among
them, dibutyl-, dipentyl-, and dihexyl-substituted polygermanes gave the best synthetic results, i.e., relatively high
yields (25±45%), high current eciency (25±50%), high molecular weights (Mw: 10,000±25,000), and long
absorption maxima (lmax: 324±327 nms). Thus obtained polygermanes showed the discontinuous thermochromic
changes known for the polymers. Syntheses of the corresponding polysilanes under the same conditions gave better
yields (50±70%), better current eciency (50±80%), slightly larger molecular weights (Mw: 15,000±35,000), but
shorter absorption maxima (lmax: 313±316 nms). The reaction mechanism was discussed in detail. It was suggested
in the discussion that more back-biting reactions, which is one of the causes for the decrease of the yield and the
molecular weight, occur in the later stages of electrolysis. In order to know if molecular weights and yields are
decreasing in the later stages of electropolymerization, syntheses with 40, 60, 80, and 100% electricity of the
theoretical value were carried out. Synthesis with 80% electricity of the theoretical value gave the best results.
# 1998 Elsevier Science Ltd. All rights reserved.
Keywords: s-conjugated polymers; Polygermane; Polysilane; Electropolymerization; Electrosynthesis
1. Introduction
ring-opening polymerization of cyclosilanes [18], and
anionic polymerization of masked disilenes [19].
Among them the most often employed synthetic path-
way is the Wurtz coupling reaction of diorganodihalo-
geno-silanes and -germanes with sodium metal in
boiling toluene [3, 4, 8, 9]. However, because of the
dangerous reaction conditions due to the use of alkali
metals in Wurtz coupling reaction, electrochemical re-
Polysilanes have been intensely investigated as a new
class of functional materials such as silicon carbide
precursor, photoresist, and photoinitiator [1, 2].
Polygermanes are analogues of polysilanes [3, 4]. They
are known to show properties similar to those of the
corresponding polysilanes, such as thermochromic
properties [5], nonlinear optical properties [6], etc. [7].
These polymers are now known as s-conjugated poly-
mers because these polymers have delocalized s-elec-
trons along their main chains.
duction attracted attention as
pathway [10±16].
a safer synthetic
Electrochemical formation of s-bonding in s-conju-
gated compounds was ®rstly achieved by Hengge et al.
in 1976 [20]. They obtained hexamethyldisilane by the
reduction of trimethylchlorosilane. Since then, many
researchers have reported electrochemical syntheses of
polysilane oligomers and polymers [10±16, 21±24].
Employment of mercury as a `sacri®cial' anode in the
syntheses of polysilane oligomers was originally
reported by Hengge et al., but the yields were not
high [20, 21]. Later by Kunai et al., employment of
undivided cell and examination of experimental con-
There are several synthetic pathways to these poly-
mers, i.e., the Wurtz coupling reaction [3, 4, 8, 9] and
electrochemical reduction [10±16] of dihalogeno-ger-
manes and -silanes, dehydrogenative coupling reaction
of hydrosilanes using transition metal catalysts [17],
* Corresponding author. Fax: +81-462423000; e-mail: oka-
no@photo.t-kougei.ac.jp
0013-4686/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved.
PII: S0013-4686(98)00173-X

660
M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
various oligomer
ditions
were
done
for
economical industrial production. The cathode ma-
terial was platinum which is most often used in electro-
chemistry. A platinum plate (1 cm  4 cm) was used
for the cathode. Silver was chosen for the anode ma-
terial because silver swiftly traps halogens and the
resulting silver halogenide does not prevent electroche-
mical reactions. Mercury was also a candidate for the
anode material. However, it was not suitable because
in preliminary experiments, considerable amount of
mercury deposited on the cathode. The supporting
electrolyte was TBAP and the solvent was 1,2-
dimethoxyethane (DME). Tetrahydrofuran (THF) was
also a candidate for the solvent. However, in electro-
polymerization of dibutyldichlorogermane, large
amount of solvent was incorporated in the resulting
(Bu2Ge)n as copolymerized tetramethylene oxide1.
syntheses [12, 22, 25]. Large improvement of yields and
examination of silver, aluminum, and copper anodes
were performed [12, 22]. Shono et al. reported synth-
eses of polysilanes with Mg electrodes. Their method
gave fairly good yields and molecular weights, how-
ever, required sonication or the alternation of the cath-
ode and the anode. It also spent a lot of electricity
1
typically 6.4±8.6 F mol [11, 13].
Reports on the electrochemical syntheses of polyger-
mane oligomers and polymers are much more limited
than those concerning polysilanes. Preliminary results
were reported by Shono et al. [13] and by Aeiyach et
al. [14] using Mg electrodes, and by Okano et al. [15] using
Pt cathode and Ag anode. Recently, systematic study on
the electrodes were carried out for polydihexylgermane by
Martins et al. [16]. However, systematic study on electro-
chemical syntheses of various polygermanes and charac-
teristics of them, have never been reported.
2.3. Electropolymerization and isolation
The purpose of this paper is (1) to examine the val-
idity of our previously reported method [15], as an
example of electrochemical method, for preparation of
various polygermanes, (2) to compare electrochemical
syntheses of polygermanes with those of polysilanes,
and (3) to give better image on the reaction mechanism
of electrochemical synthesis.
All procedures were carried out in the dark. 1.95 g
(5.7 mmol) of TBAP, previously dried under vacuum
for more than 6 h, was placed in the electrochemical
cell, and dried again in the cell in vacuum for 1 h.
After ®lling the cell with dry N2, 30 ml of DME and
subsequently typically 15±20 mmol of the starting
compound (dichlorogermane unless otherwise noted)
were introduced into the cell through its sampling port
equipped with a rubber syringe cap. The amount of
the introduced starting compound was calculated from
the weight increase of the entire cell assembly.
Electrolysis was carried out at room temperature at
constant voltage of 20 V while the electrolyte solution
was stirred magnetically. 25 to 50 mA was the initial
current and it decreased gradually as the reaction pro-
ceeded. Charge passing through the circuit was moni-
tored by a coulomb meter. Electrolyses was terminated
at a desired electricity calculated prior to electrolysis
based on Eq. (1) and the amount of the starting com-
pound weighed above.
2. Experimental
2.1. Materials
Dichlorogermanes and dichlorosilanes were the
monomers unless noted as dibromogermanes.
Tetrabutylammonium perchlorate (TBAP) was precipi-
tated in pentane±ethyl acetate solution and dried in
vacuum before use. 1,2-Dimethoxyethane was dried
with calcium hydride and distilled under nitrogen at-
mosphere.
2.2. Electrochemical set-up
‡2ne
nR1R2GeX2 4ꢀR1R2Ge†n ‡ 2nX
ꢀ1†
A one-compartment cell equipped with a vacuum
valve and a sampling port was employed for the elec-
trolyses. Our electrode system was optimized to study
the polymerization mechanism, but not to carry out
R=Me, Et, Pr, Bu, Pen, Hex, Ph
X=Cl, Br
After electrolysis, the resulting solution was concen-
trated to half of the original volume under a reduced
pressure, and poured into 300 ml of methanol. The
polymer precipitated in methanol (TBAP was dissolved
in methanol and separated from the polymer). Then,
the polymer was collected and redissolved in 50 ml of
toluene. The toluene solution was poured into 300 ml
of methanol. Such dissolution and precipitation pro-
cedure was repeated three times. When a trace of silver
was observed in the polymer, a pure sample was
obtained by Soxhlet extraction method using toluene
1 In THF, the authors tried polymerization of Bu2GeCl2,
Pen2GeCl2, Hex2GeCl2, PhMeGeCl2, Bu2SiCl2, Pen2SiCl2,
and Hex2SiCl2. Polymerization of Bu2GeCl2 was the only case
which showed copolymerization of tetramethylene oxide.
With the other tried monomers, electropolymerization in THF
gave slightly higher yields and molecular weights compared
with those in DME. Martins et al. also reported formation of
the polytetrahydrofuran in their electrochemical system for
polymerization of (Hex2Ge)n in Ref. [16].

M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
661
as the solvent. Finally, the collected polymer was dried
in vacuum.
in toluene. Fractions were obtained by stepwise ad-
dition of methanol to the solution and collection of
the precipitated polymer.
2.4. Characterization of polymers
3. Results and discussion
Identi®cation of polymers was carried out by com-
paring their properties with the reported values: poly(-
3.1. Syntheses of (Bu2Ge)n, (Pen2Ge)n, and (Hex2Ge)n
diethylgermylene),
poly(dipropylgermylene), (Pr2Ge)n [3], poly(dibutylger-
mylene), (Bu2Ge)n [3], poly(dipentylgermylene),
(Et2Ge)n
[3],
Typical results are summarized in Table 1. Syntheses
of (Bu2Ge)n, (Pen2Ge)n, and (Hex2Ge)n, gave the best
results, i.e., relatively high yields, high molecular
weights, and long absorption max's among the poly-
germanes examined in this research. These polymers
precipitated as white powder in methanol. Usually,
yields and current eciencies are 20±45% and 20±
50%, respectively, Mw's are 10,000 to 25,000, lmax's
are 324 to 327 nms.
The values of lmax's clearly indicate the uninter-
rupted catenation of germanium atoms, that is, for-
mation of polygermanes. 1H NMR and IR spectra of
these polymers agreed well with those reported in the
literature [3, 5]. A UV spectrum of (Hex2Ge)n is shown
in Fig. 1 as an example. The shape of the spectrum
very closely resembled that reported for (Hex2Ge)n [3]
and also that reported for (Hex2Si)n [32]. All the three
polymers showed so called discontinuous thermochro-
mic changes [1]. That of the obtained (Hex2Ge)n is
shown in Fig. 2. There is no experimental observation
which indicates any di€erence between the polyger-
manes obtained by electrolysis and those obtained by
Wurtz coupling reaction.
(Pen2Ge)n [5], poly(dihexylgermylene), (Hex2Ge)n [3],
poly(methylphenylgermylene), (PhMeGe)n [3], poly(di-
butylsilylene), (Bu2Si)n [26], poly(dipentylsilylene),
(Pen2Si)n [26], poly(dihexylsilylene), (Hex2Si)n [26, 27],
and poly(methylphenylsilylene), (PhMeSi)n [28, 29].
Identi®cation of (Bu2Si)n and (Pen2Si)n was also done
by comparing their properties with those of samples
synthesized by an authentic method [30, 31].
Poly(butylphenylgermylene), (PhBuGe)n, and poly(di-
phenylgermylene), (Ph2Ge)n were identi®ed by UV, IR,
and 1H NMR (not measured for (Ph2Ge)n because it
has very low solubility to common solvents). UV spec-
tra were measured in tetrahydrofuran unless otherwise
noted because tetrahydrofuran dissolved wide variety
of
polygermanes
and
polysilanes
well.
Thermochromism was measured in pentane because
pentane enables measurements down to 1208C. 1H
NMR spectra were measured in CDCl3. IR spectra
were measured by KBr method. Molecular weights
were determined by gel-permeation chromatography
relative to polystyrene standards. The wavelength of
the UV detector was set at 254 nm. Fractionation of
polymers was carried out as follows when necessary.
An electrochemically prepared polymer was dissolved
Yields were slightly higher than the values reported
for Wurtz coupling reactions by alkali metals [3].
Molecular weight distribution was polymodal and mol-
Table 1
Synthetic results for the electrochemical polymerization of germanes and silanes
Yield (%)a
lmax (nm)b
Mw (Â103)c
Mn (Â103)c
Mw/Mn
c
Polymer
Monomer
(Et2Ge)n
(Pr2Ge)n
Et2GeCl2
Pr2GeCl2
18
6.8
31
20
41
11
16
25
62
56
68
14
293
0.86
1.6
14
0.22
1.3
2.0
3.9
6.1
1.7
1.6
4.0
1.2
7.1
2.8
1.7
5.7
2.5
312
325
327
325
327
337
Ðe
(Bu2Ge)n
(Pen2Ge)n
(Hex2Ge)n
(PhMeGe)n
(PhBuGe)n
(Ph2Ge)dn
(Bu2Si)n
Bu2GeCl2
Pen2GeCl2
Hex2GeCl2
PhMeGeCl2
PhBuGeCl2
Ph2GeBr2
Bu2SiCl2
11
10
9.4
3.9
Ðe
Ðe
Ðe
314
315
314
334
16
32
19
8.5
6.8
4.1
3.0
1.9
4.7
4.6
2.4
(Pen2Si)n
(Hex2Si)n
(PhMeSi)n
Pen2SiCl2
Hex2SiCl2
PhMeSiCl2
7.3
a Current yields are the same values.
b As isolated (see also Mw and Mn), in THF.
c As isolated.
d Not identi®ed, probably contains oligomers.
e Not soluble in common solvents.

662
M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
the lower molecular weight of polygermanes compared
to the corresponding polysilanes are discussed later.
3.2. Syntheses of (Et2Ge)n and (Pr2Ge)n
Syntheses of (Et2Ge)n and (Pr2Ge)n gave low yields of
less than 20%, low current eciencies less than 25%,
low molecular weights under 1.5 Â 103, and short
absorption max's shorter than 320 nm. The polymers
were obtained as a slightly pasty white solid. The value
of lmax's indicate that these polymers have shorter ger-
manium chain compared to the (Bu2Ge)n, (Pen2Ge)n,
and (Hex2Ge)n. Though the molecular weights deter-
mined based upon polystyrene standards suggest 6 to 7
germanium catenation, the values of lmax's more clearly
show that germanium catenations are longer than the
values suggested by molecular weights, because lmax of
Me(Me2Ge)6Me is only 255.3 nm [33]. It should be
noted that molecular weight distributions were deter-
mined using the UV detector set at 254 nm.
Measurements at longer wavelengths give larger Mw
and Mn values. 1H NMR and IR spectra of the obtained
polymers agreed well with those reported in the
literature [3]. The shapes of UV spectra of (Et2Ge)n and
(Pr2Ge)n were principally the same as those observed for
(Bu2Ge)n, (Pen2Ge)n, and (Hex2Ge)n. As expected from
the previous reports [3], their thermochromic behavior
was not the discontinuous changes but gradual red shifts
as the temperature was lowered.
Fig. 1. Absorption spectrum of (Hex2Ge)n (Mw=29,950,
Mn=28,556, Mw/Mn=1.05).
ecular weight was widely distributed. Mw/Mn values
were usually 2 to 4 or larger.
Results for the polysilanes with the same substituent
groups are also listed in Table 1. Electro-
polymerization of the corresponding polysilanes under
the same conditions gave better yields (55±70%), better
current eciencies (50±80%), and larger molecular
weights (15,000±35,000). In Wurtz coupling reactions,
it is also true that polysilanes give much better results.
Molecular weights above 105 have been reported while
they have never exceeded 20,000 in polygermane synth-
eses. The reason(s) for (1) the lower molecular weight
of poly-germanes and -silanes in electrochemical synth-
eses compared to Wurtz coupling syntheses, and (2)
3.3. Syntheses of (Ph2Ge)n, (PhMeGe)n and
(PhBuGe)n
Poly(diphenylsilane), (Ph2Si)n, is well known to have
very low solubilities in common solvents [34].
Electrolysis of Ph2GeBr2 also gave white solid not sol-
uble in common solvents. The authors believe that the
product contained poly(diphenylgermane), (Ph2Ge)n,
molecules as well as 5 or 6 membered cyclics, because
the measurement of a di€use re¯ection spectrum of the
product showed absorption peaks at 320 and
380(shoulder) nms. It is not likely that absorption
peaks of small cyclic compounds reach 380 nm. IR
data for (Ph2Ge)n are as follows. IR (KBr): 2850±3080,
1580, 1480, 1430, 1380, 1330, 1300, 1260, 1190, 1160,
1
1080, 1020, 1000, 970, 910, 740, 700, 620 cm
.
Syntheses of (PhMeGe)n and (PhBuGe)n gave soluble
polymers. 1H NMR and IR spectra of the (PhMeGe)n
agreed well with those reported in the literature [3]. Its
UV spectrum had a small peak at 269 nm besides its
main peak at 327 nm. Such small peaks have been
known to exist in absorption spectra of phenyl substi-
tuted polysilanes [35]. Mw of (PhMeGe)n was 6000±
8000, which was signi®cantly smaller than those for
(Bu2Ge)n, (Pen2Ge)n, (Hex2Ge)n. Such lower molecular
weights of phenyl-substituted polymers have also been
Fig. 2. Thermochromic behavior of the electrochemically pre-
pared (Hex2Ge)n (Mw=29,950, Mn=28,556, Mw/Mn=1.05).

M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
observed for the polysilanes obtained by the Wurtz
663
coupling method (190,000 for (PhMeSi)n while
1,800,000 for (Bu2Si)n) [1]. The authors repeated polym-
erization of (PhMeGe)n carefully, but obtained no better
results under the experimental conditions. Employment
of PhMeGeBr2 also gave no improvements.
Electropolymerization of (PhMeSi)n was similar to elec-
tropolymerization of (PhMeGe)n in giving lower yields
and lower molecular weights compared to those for elec-
trochemically synthesized dibutyl, dipentyl, and dihexyl
substituted polysilanes.
Yields and molecular weights of (PhBuGe)n were also
low. The reason for such poor results of phenyl substi-
tuted polymers may be their ability to easily accept elec-
trons from the cathode. Acceptance of electrons and
cleavage of Si±Si or Ge±Ge bonds are well known [36].
(See also the following discussion on the reaction mech-
1
anism.) H NMR, UV and IR data for (PhBuGe)n are
as follows. 1H NMR (CDCl3): d 0.77(br m); 7.00 (br m).
UV (THF): 269 and 337 nm. IR (KBr): 2800±3100,
1580, 1460, 1430, 1380, 1340, 1300, 1260, 1180, 1080,
1
1020, 970, 870, 800, 730, 700, 630 cm
.
3.4. On the reaction mechanism
Reaction mechanism for electrochemical synthesis of
s-conjugated polymers has scarcely been discussed in
the literature. Plausible reactions are listed below (Eq.
2-1±Eq. 2-11).
Eq. 2-1 to Eq. 2-3 are the reactions expected to
occur on the platinum cathode. Eq. 2-1 is the re-
duction of monomer and Eqs. 2-2 and 2-3 are the re-
duction of polymeric compounds. In an experiment,
the consumption of monomer molecules was moni-
tored by gas chromatography. Most of monomer mol-
ecules were consumed at the electrolysis with 50% of
the theoretically calculated electricity. So, it was con-
®rmed that monomers are swiftly consumed in the
early stage of electrolysis.
It has already been proved by the authors in a sep-
arate work that electroreduction of s-conjugated com-
pounds makes them cleave into anions (Eqs. 2-2 or 2-
3) [36]. The authors believe that Eq. 2-3 is preferred to
Eq. 2-2, but still can not exclude the possibility of Eq.
2-2. Of course, in the early stage of electrolysis, Eq. 2-
1 is the dominant reaction, and in the ®nal stage of
electrolysis, Eq. 2-3 is the dominant reaction on the
electrode.
Eq. 2-4 to Eq. 2-11 are the reactions of anionic
species with neutral compounds expected to occur in
the solution. In Wurtz coupling reactions, these reac-
tions occur only on the metal surface and in the vicin-
ity of the metal surface because such anions can not
survive in bulk toluene solution. However, not like
those in Wurtz coupling reactions in toluene, these
reactions play important roles in electrochemical synth-

664
M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
Table 2
In¯uence of electricity on the electropolymerization of (Bu2Ge)n
d
Electricity (%)
Yield (g)/(%)
C.E.b
lmax (nm)c
Mw (Â103)d
Mn (Â103)d
Mw/Mn
40
60
0.47/17
1.09/37
1.19/43
0.89j/31
40
62
54
40
326
11
12
23
15
2.7
3.1
4.1
2.0
4.1
3.9
5.6
7.5
325
327
325
80
100a
1
a Corresponds to 2 F mol
b Current eciency.
.
c As isolated (see also Mw and Mn), in THF.
d As isolated.
eses, because anionic species are much more stable in
polar solvent such as 1,2-dimethoxyethane than in
toluene. They will strongly in¯uence the yields and the
molecular weights. It has been reported for Wurtz
coupling reactions that the molecular weights of the
resulting polymers were smaller when they were poly-
merized in polar solvents such as THF ( [2] and refer-
ences cited therein).
3.5. Discussion on the yield and the molecular weight
Yields are discussed ®rst. Since polymers were preci-
pitated in toluene±methanol solution, the missing ger-
manium(silicon) compounds had to be dissolved in the
solution. Short chain oligomers and small cyclic com-
pounds were expected to be dissolved. Such com-
pounds were mainly produced by Eq. 2-3 on the
electrode and Eq. 2-8 (back-biting) in the solution.
(The authors believe that formation of other com-
pounds, such as those which had lost one of their sub-
stituent groups and those with reacted substituent
groups, are negligible.) In order to improve the yields,
suppression of the reactions (Eqs. 2-3 and 2-8) seems
essential. If there is some way to isolate the dissolved
oligomers and cyclics, they may be reacted again to
form longer polymers. Such recycling of compounds is
especially important in the syntheses of polygermanes
because germanium compounds are much more ex-
pensive than silicon compounds.
Eq. 2-4 to Eq. 2-6 are the reactions of monomeric
anions. Eq. 2-4 is the reaction of monomeric anion
with monomers and is the dominant reaction in the
very beginning of electrolysis. Eqs. 2-5 and 2-6 are the
reactions of monomeric anions with oligo(poly)mers.
Both reactions are expected to progress smoothly.
However, longer the oligo(poly)mer, lower the chances
of reaction Eq. 2-5.
Eq. 2-7 to Eq. 2-11 are the reactions of polymeric
anions. Eqs. 2-7 and 2-8 are the reactions in the mol-
ecule themselves. Eq. 2-7 is an end-biting reaction and
Eq. 2-8 is an back-biting reaction. In polygermane and
polysilane syntheses, the word `back-biting' has been
used to indicate Eq. 2-8 which emits small cyclic (typi-
cally 5 to 6 membered rings) compounds [9, 37±39].
However, emission of such small cyclic compounds
should not be too much emphasized. Large cyclics
should be produced as well. The authors have been
also carrying out an investigation focusing on the
back-biting phenomena in electrochemical systems [40].
It was concluded in the work that back-biting reac-
tions occur in the course of electropolymerization of s-
conjugated polymers.
Nextly, the molecular weights are discussed. Eqs. 2-3
and 2-8 also cause the decrease in molecular weights.
In electrochemical syntheses of polysilanes, regardless
of the electrolysis system employed for the synthesis,
molecular weights were smaller than those obtained by
Wurtz coupling reactions. The fact for the polyger-
mane syntheses that here reported molecular weights
are larger than those reported for the Wurtz coupling
reactions, may sound inconsistent with the polysilane
syntheses. However, it should be noticed the molecular
weights of polygermanes obtained by the Wurtz coup-
ling reactions hardly exceed 10,000 while those of poly-
silanes exceed 100,000. The surprisingly low molecular
weights of polygermanes by the Wurtz coupling reac-
tion is the cause for the apparent inconsistency.
Eq. 2-9 to Eq. 2-11 are the reactions of polymeric
anions with neutral compounds. Eq. 2-9 should scar-
cely occur because monomers are consumed swiftly
before appearance of polymers of signi®cant lengths.
Eqs. 2-10 and 2-11 are the intermolecular versions of
Eqs. 2-5 and 2-6, respectively. Eq. 2-11 should be
named as side-biting reaction instead of back-biting
reaction. As discussed above, Eq. 2-11 is more likely
than Eq. 2-10.
In electrochemical syntheses of polygermanes, mol-
ecular weights were smaller than those obtained for
the corresponding polysilanes. As pointed out above,
such tendency is more clearly observed in the syntheses
by the Wurtz coupling reactions. As far as the authors
concern, there is no experimental indication which
suggests the di€erence between polygermanes and
polysilanes in ease of the chain elongation reactions

M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
(Eqs. 2-4, 2-5, (2-6), 2-9, 2-10, and (2-11)). Thus, the 4. Conclusion
665
chain degradation reactions (Eqs. 2-3 and 2-8) seem re-
sponsible for the relatively low molecular weight of
polygermanes2. Since polygermanes have longer lmax's
than the corresponding polysilanes of the same mol-
ecular weight, it is suggested that polygermanes accept
electrons more easily on the electrode (Eq. 2-3).
Moreover, since germanium±germanium bonds are
slightly longer than silicon±silicon bonds, it is likely
that germanium±germanium bonds more easily be
attacked by anions (Eq. 2-8).
Five dialkyl-substituted and three phenyl-substituted
polygermanes were polymerized electrochemically.
Dibutyl-, dipentyl-, and dihexyl-substituted polyger-
manes gave the best synthetic results with relatively
high yields, high current eciencies, high molecular
weights, and long absorption maxima. Syntheses of the
corresponding polysilanes under the same conditions
gave better yields, better current eciencies, slightly
larger molecular weights, but shorter absorption max-
ima. Syntheses of polygermanes with shorter alkyl-sub-
stituents gave relatively low yields, low current
eciencies, low molecular weights, and short absorp-
tion maxima. Phenyl-substituted polygermanes gave
relatively low yields, low current eciencies, low mol-
ecular weights, but long absorption maxima.
Electropolymerization with electricity 20% less than
the theoretical value, gave the best synthetic results.
The discussion on the yield and the molecular weight
revealed that the reactions, (1) cleavage of metal±metal
bond on the electrode and (2) back-biting reaction in
the solution, should be suppressed in order to obtain
better synthetic results.
3.6. Examination of electricity for the
electropolymerization
It was suggested above that suppression of Eqs. 2-3
and 2-8 is important to obtain better synthetic results.
The reactions are expected to occur more in the con-
ditions met in the later stages of electrolysis. For that
reason, electrolyses with less electricity, 20, 40, 60, and
80% of the theoretically calculated value were carried
out. Results for (Bu2Ge)n are summarized in Table 2.
Electropolymerization, carried out with electricity 80%
of the theoretical value, gave the best yield, molecular
weight, and absorption maximum. Such improvement
of synthetic results with less electricity, was observed
for all the dialkyl-substituted polygermanes and polysi-
lanes examined here, though it was not obvious in
cases of (Et2Ge)n and (Pr2Ge)n.
No di€erence was found between polygermanes syn-
thesized by Wurtz coupling reactions and those syn-
thesized here electrochemically.
3.7. Status of platinum cathode
References
[1] R. West, J. Organomet. Chem. 300 (1986) 327.
[2] R.D. Miller, J. Michl, Chem. Rev. 89 (1989) 1359.
[3] K. Mochida, H. Chiba, J. Organomet. Chem. 473 (1994)
45.
As well studied by Martins et al., selection of the
cathode material is very important [16]. In our electro-
chemical system, the platinum cathode may be covered
by a thin silver layer. Trace of silver was often
observed in the raw polymer product. In order to
obtain a better image on our cathode surface, ¯uor-
escent X-ray analysis was done. The ratio of silver to
platinum was 0.5±1.0. A signi®cant amount of silver
was found on platinum. Also, electrochemical exper-
iments were carried out using (1) silver-free new plati-
num plate and (2) silver plate cathode. However, both
experiments gave no signi®cant di€erences in yield,
[4] K. Mochida, Nippon Kagaku Kaishi (1994) 224.
[5] R.D. Miller, R. Sooriyakumaran, J. Polym. Sci. Polym.
Chem. 25 (1987) 111.
[6] J.-C. Baumert, G.C. Bjorklund, D.H. Jundt, M.C.
Looser, H. Looser, R.D. Miller, J. Rabolt, R.
Swalen, J.D. Swalen, R.J. Twieg, Appl. Phys. Lett. 53
(1988) 1147.
[7] K. Mochida, K. Kimijima, H. Chiba, M. Wakasa, H.
Hayashi, Organometallics 13 (1994) 404.
[8] R.E. Trujillo, J. Organomet. Chem. 198 (1980) C27.
[9] K. Matyjaszewski, D. Greszta, J.S. Hrkach, H.K. Kim,
Macromolecules 28 (1995) 59.
current eciency, molecular weight, and lmax
.
Therefore, it is likely that the signi®cant area of plati-
num cathode surface is covered with silver in the early
stage of electrolysis.
[10] C. Biran, M. Bordeau, P. Pons, M.-P. Leger, J.
Dunogues, J. Organomet. Chem. 382 (1990) C17.
[11] T. Shono, S. Kashimura, M. Ishifune, R. Nishida, J.
Chem. Soc. Chem. Commun. (1990) 1160.
2 Degradation by UV light and thermal energy are negli-
gible here because UV light is shut out throughout experimen-
tal procedures, i.e., polymerization, isolation, and analysis
and also because (Hex2Ge)n was found to be stable in toluene
solution at 1008C, though polygermanes seems to degrade fas-
ter than polysilanes by UV light [41] and by heat [42, 43].
[12] A. Kunai, T. Kawakami, E. Toyoda, M. Ishikawa,
Organometallics 10 (1991) 2001.
[13] T. Shono, S. Kashimura, H. Murase, J. Chem. Soc.
Chem. Commun. (1992) 896.
[14] S. Aeiyach, P.-C. Lacaze, J. Satge, G. Rima, Synth. Met.
58 (1993) 267.

666
M. Okano et al. / Electrochimica Acta 44 (1998) 659±666
[15] M. Okano, T. Toriumi, K. Takeda, Y. Kurimoto, T.
Hamano, H. Hamano, Denki Kagaku 62 (1994) 1163
(CAS 122:199517q).
[29] P. Trefonas III, P.I. Djurovich, X.-H. Zhang, R. West,
R.D. Miller, D. Hofer, J. Polym. Sci. Polym. Lett. Ed.
21 (1983) 819.
[16] L. Martins, S. Aeiyach, M. Jouini, P.-C. Lacaze, J.
Rima, G. Rima, Appl. Organomet. Chem. 11 (1997) 583.
[17] T.D. Tilley, Acc. Chem. Res. 26 (1993) 22.
[18] K. Matyjaszewski, Y. Fupta, M. Cypryk, J. Am. Chem.
Soc. 113 (1991) 1046.
[30] J.M. Zeigler, Polym. Prepr. (Am.Chem. Soc., Div.
Polym. Chem.) 27 (1986) 109.
[31] J.M. Zeigler, L.A. Harrah, A.W. Johnson, Polym. Prepr.
(Am. Chem. Soc., Div. Polym. Chem.) 28 (1987) 424.
[32] P. Trefonas III, J.R. Damewood Jr.., R. West, R.D.
Miller, Organometallics 4 (1985) 1318.
[19] K. Sakamoto, K. Obata, J. Hirata, M. Nakajima, H.
Sakurai, J. Am. Chem. Soc. 111 (1989) 7641.
[20] E. Hengge, G. Litscher, Angew. Chem. Int. Ed. Engl. 15
(1976) 370.
[33] M. Okano, K. Mochida, Chem. Lett. (1990) 701.
[34] F.S. Kipping, J. Chem. Soc. 125 (1924) 2291.
[35] L.A. Harrah, J.M. Ziegler, Macromolecules 20 (1987)
601.
[21] E. Hengge, H. Firgo, J. Organomet. Chem. 212 (1981)
155.
[36] T. Kugita, D. Ohtani, M. Okano, K. Mochida, Denki
Kagaku 62 (1994) 522 (CAS 121:120044f).
[22] A. Kunai, T. Kawakami, E. Toyoda, M. Ishikawa,
Organometallics 10 (1991) 893.
[37] H.K. Kim, K. Matyjaszewski, J. Polym. Sci. Polym.
Chem. 31 (1993) 299.
[23] Y. Kimata, H. Suzuki, S. Satoh, A. Kuriyama, Chem.
Lett. (1994) 1163.
[38] R.G. Jones, R.E. Ben®eld, R.H. Cragg, A.C. Swain, J.
Chem. Soc. Chem. Commun. (1992) 112.
[24] Y. Kimata, H. Suzuki, S. Satoh, A. Kuriyama,
Organometallics 14 (1995) 2506.
[39] S. Gauthier, D.J. Worsfold, Macromolecules 22 (1989)
2213.
[25] A. Kunai, E. Toyoda, T. Kawakami, M. Ishikawa,
Organometallics 11 (1992) 2899.
[40] M. Okano, D. Ohtani, H. Hamano, Denki Kagaku 64
(1996) 1203 (CAS 125:301766w).
[26] R. Menescal, J. Eveland, R. West, M. Blazso,
Macromolecules 27 (1994) 5893.
[41] K. Mochida, Main Group Metal Chem. 17 (1994) 25.
[42] B. Jambe, J. Marchand-Brynaert, J. Devaux, J. Polym.
Sci. Polym. Chem. 33 (1995) 1283.
[27] J.F. Rabolt, D. Hofer, R.D. Miller, G.N. Fickes,
Macromolecules 19 (1986) 611.
[28] R.E. Trujillo, J. Organomet. Chem. 198 (1980) C27.
[43] R.A. Jackson, J. Organomet. Chem. 166 (1979) 17.